Re: [Qemu-devel] [PATCH] linux-user: Enable sigaltstack syscall for sh4

[Laurent Vivier <lvivier@redhat.com>]

On 19/11/2015 12:15, Peter Maydell wrote:
> On 19 November 2015 at 10:22, Laurent Vivier <lvivier@redhat.com> wrote:
>> Hi,
>>
>> On 19/11/2015 10:28, John Paul Adrian Glaubitz wrote:
>>> On 11/19/2015 10:17 AM, Peter Maydell wrote:
>>>> Unfortunately this isn't sufficient. You also need to add
>>>> the code to the sh4-specific functions in linux-user/signal.c
>>>> which honours the requested sigaltstack when taking and returning
>>>> from signal handlers.
>>
>> it seems all needed functions for sh4 signal handling are already
>> written in linux-user/signal.c, I thing about setup_frame(),
>> setup_rt_frame(), do_sigreturn() and do_rt_sigreturn().
> 
> Other architectures have calls to do_sigaltstack or other
> sigaltstack handling in their frame setup and sigreturn
> functions. SH4 doesn't, which implied to me that we were
> missing something beyond just enabling the call to do_sigaltstack.
> Maybe that's for something else?

Well, I didn't really check the code in these functions, but SH4 has the
call in do_rt_sigreturn().

The only check I did is to compare setup_frame() with the one from
kernel and it seems ok.

But I think the patch from Adrian is just fixing a forgetting in the
original SH4 patch:

c3b5bc8 SH4: Signal handling for the user space emulator, by Magnus Damm.

This patch uses do_sigaltstack() without enabling sigaltstack() syscall.

> (Incidentally the only targets other than SH4 not listed in
> this ifdef are CRIS, Microblaze, TileGX and Unicore32.
> Maybe we should just go ahead and remove the ifdef entirely,
> if do_sigaltstack() doesn't really have any target-specific
> requirements?)

I don't know if these targets have a lot of users. At least for SH4, we
can enable it and see if someone is screaming (I like the "crash-test
method" :) ). I think Adrian will take care of this (and I can help).

Thanks,
Laurent